Revit ortamında inşaat sahası izleme amaçlı GPS ve UWB verilerinin sorunsuz entegrasyonu

This thesis proposes a hybrid real-time location tracking framework designed for construction site safety, which integrates Global Positioning System (GPS) and Ultra-Wideband (UWB) data using an Unscented Kalman Filter (UKF). The fused positions are dynamically mapped into a Building Information Modeling (BIM) environment via the Autodesk Revit API and visualized using custom Python scripts in pyRevit. The experimental setup involved both indoor and outdoor test areas across two buildings. UWB anchors were deployed in fixed tetrahedral configurations indoors, while test subjects wore mobile tags to simulate worker movement. Outdoor position data was collected via iOS-based GPS smartphone applications. The data was post-processed and synchronized, then filtered using Mahalanobis distance to remove 3D outliers before applying UKF-based fusion.Transformed coordinates were visualized in Revit, enabling tracking and visulization of the workers. The experimental results revealed critical limitations in the accuracy of the iOS-based GPS data. Positional deviations in open outdoor areas frequently exceeded 6 meters and reached nearly 10 meters in obstructed environments such as near buildings or trees. The study recommends adopting GNSS receivers such as dual-frequency RTK systems or integrating inertial navigation sensors to enhance outdoor data fidelity. Despite the shortcomings of GPS, the UWB-based indoor localization performed well under line-of-sight (LOS) conditions. However, the study also identified significant performance degradation in Non-Line-of-Sight (NLOS) scenarios. Overall, the research demonstrates that the proposed GPS–UWB–UKF fusion system significantly improves positional continuity and spatial awareness when integrated in Revit. This study contributes an extendable methodology for real-time worker safety management based on sensor fusion, spatial filtering, and Revit integration while simultaneously underscoring the practical hardware and calibration limitations in mobile sensor systems.